Our recent discovery, using a whole-genome association mapping approach on 96 cases and 50 controls, of strong association between a variant in complement factor H, CFH Y402H, and AMD has been shown in 10 different Caucasian populations. Across these 10 cohorts, frequencies between cases and controls are strikingly consistent: They show that controls are more likely but not exclusively to have YY (42% vs. 18%) and cases are more likely to be HH (35% vs. 13%). The estimated genetic risk does not explain the entire risk profile conferring the disease. Therefore, it is reasonable to assume that there are additional genetic and/or environmental factors acting independently or in concert in the development of AMD. The questions we would like to resolve here are: What are the risk factors that drive those 18% with two copies of non-risk CFH alleles to AMD, and what protects those 13% who carry two risk alleles from having AMD? To address these questions, we identified homozygous AMD/control individuals in AREDS who carried two copies of histidine (H) or two copies of tyrosine (Y) of the CFH 402 variant. DNA samples of homozygous CFH402 individuals will be genotyped using a whole-genome SNP microarray platform at a resolution of 317,000 tag SNPs derived from the HapMap data. Statistical and bioinformatics methods/algorithms will be developed, and analyses will be performed on the resulting genotype and phenotype data with three specific aims: 1. To discover the genetic/environmental risk factors and their interactions in AMD patients who do not carry the disease risk allele of complement factor H, i.e., AMD with CFH 402 YY; 2. To discover the protective factors in individuals who do not present with AMD but carry two copies of the CFH risk alleles, i.e., None-AMD with CFH 402 HH; 3. To discover the epistatic genetic variant(s) in AMD patients with two copies of CFH 402 risk alleles, i.e., AMD with CFH 402 HH. [unreadable] [unreadable] [unreadable] [unreadable]